key: cord-0809029-4vgyx58i
authors: Østergaard, Lauge; Butt, Jawad Haider; Kragholm, Kristian; Schou, Morten; Phelps, Matthew; Sørensen, Rikke; Lamberts, Morten; Gislason, Gunnar; Torp-Pedersen, Christian; Køber, Lars; Fosbøl, Emil L.
title: Incidence of acute coronary syndrome during national lock-down: insights from nationwide data during the coronavirus disease 2019 (COVID-19) pandemic
date: 2020-11-06
journal: Am Heart J
DOI: 10.1016/j.ahj.2020.11.004
sha: 5b5ffc5334ff0597e1bd0e271c6647a25ce8c4e5
doc_id: 809029
cord_uid: 4vgyx58i

BACKGROUND: Urgent recognition and treatment are needed in patients with acute coronary syndrome (ACS), however this may be difficult during the Coronavirus disease 2019 (COVID-19) pandemic with a national lock-down. We aimed to examine the incidence of ACS after national lock-down. METHODS: The Danish government announced national lock-down on March 11 2020 and first phase of reopening was announced on April 6. Using Danish nationwide registries, we identified first-time ACS admissions in A) January 1 – May 7 2017-2019, and B) January 1 – May 6 2020. Incidence rates of ACS admissions per week for the 2017-2019-period and the 2020-period were computed and incidence rate ratios (IRR) were computed using Poisson regression analysis. RESULTS: The number of ACS admissions were 8,204 (34.6% female, median age 68.3 years) and 2,577 (34.0% female, median age 68.5 years) for the 2017-2019- and 2020-period, respectively. No significant differences in IRRs were identified for weeks 1-9 (January 1 – March 4) for 2020 compared with week 1-9 for 2017-2019. In 2020, significant lower IRRs were identified for week 10 (March 5-11) IRR=0.71 (95% CI: 0.58-0.87), week 11 (12-18 March) IRR=0.68 (0.56-0.84), and week 14 (2 – 8 April) IRR=0.79 (0.65-0.97). No significant differences in IRRs were identified for week 15-18 (April 9 – May 6). In subgroup analysis, we identified that the main result was driven by male patients, and patients ≥60 years. CONCLUSION: During the COVID-19 pandemic with an established national lock-down we identified a significant decline around 30% in the incidence of ACS admissions. Along with the reopening of society, ACS admissions were stabilized at levels equal to previous years.

During the global Coronavirus disease 2019 (COVID-19) pandemic international efforts have been raised to contain spreading of the coronavirus and avoid collapsing of national healthcare systems due to high disease activity and extensive need for intensive care and respirator treatment. In Denmark, the first person confirmed COVID-19-positive was identified at 27 February 2020.(1) Danish authorities announced national lock-down on 11

March by imposing closure of Danish borders, sending home all employees working in nonessential functions in the public sector, and closure of schools and daycare institutions. One week later on 18 March a ban of gatherings of more than 10 people was instated. As admissions and deaths related to COVID-19 stabilized, Danish authorities announced the first phase of reopening on 6 April 2020.

Acute coronary syndrome (ACS) may be triggered by physical, but also psychological stress (2) (3) (4) -factors that currently are being put to the test during the global spread of the COVID-19. National health care systems are being reorganized and optimized for handling patients with COVID-19 and this may also impact patients with other diseases, including ACS. Prior studies have shown that there are several ways that such epidemics have influence on other parts of the society and might induce collateral damage in patients with ACS is feared. (5) We set out to study the incidence of ACS after the national lock-down of Denmark and how ACS admission rates changed when the first phase of societal reopening was established.

All citizens in Denmark are provided with a unique identifier and linkage of nationwide health care registries is therefore possible. For the present study the following registries were included: the Danish National Patient Registry, The Civil Population Registry, and the National Prescription Registry. The National Patient Registry holds information on every hospitalization in Denmark since 1977. One primary diagnosis code is provided for every hospital admission according to the International Classification of Disease 10 (ICD-10).

Every admission may be provided with up to several secondary diagnoses. The Civil

Population Registry holds information of every Danish citizen regarding date of birth, date of death, sex, and migration status. The Prescription Registry holds information on every filled prescription from a Danish pharmacy. The registries are described in details previously. (6) (7) (8) The Prescription Registry held information up until 29 February 2020.

The Danish health care system is tax-payer funded and universal health care is provided with few exceptions (physiotherapist and dentist). On 11 March 2020, the Danish government introduced a national lock-down, sending home public employees, closing schools, and daycare institutions. On 18 March national borders were closed. The health care system was reorganized, out-patient visits were converted to telemedicine or cancelled whenever possible, elective none-vital procedures were cancelled or postponed, emergency rooms and intensive care units were upgraded. On April 6 2020 the Danish government announced the first phase of a slowly reopening of society. Primary schools reopened at April 15 2020 and small businesses reopened at April 20, 2020.

The study population comprised every Danish citizen in two periods: 1) A combined period of 1 January -7 May 2017, 1 January -7 May 2018 and 1 January -7 May 2019, 2) 1 January -6 May 2020. The total follow-up time had a maximum of 126 days for every period. Due to leap year in 2020 it was required that end dates differed between study periods. People were required to be alive and over 18 years at index (1 January 2017-2020, respectively). People who had been admitted to hospital with ACS prior to index were excluded. has been validated with a PPV of 97%.(9) It was required that length of hospital stay was at least 24 hours, unless patients died within 24 hours of admission. Figure 1 shows a flow chart of the patient selection. Every person included in the study cohort was followed from date of entry until: ACS admission, date of death, or date of exit (7 May 2017-2019, or 6 May 2020, respectively), whichever came first.

Demographic data (age and sex) were assessed from the Civil Population Registry.

Comorbidities were registered from the National Patient Registry as an in-patient or outpatient, primary or secondary diagnosis code at any time prior to ACS admission. The following comorbidities were assessed: stroke, atrial fibrillation, peripheral vascular disease, heart failure, chronic renal failure, acute renal failure, diabetes, chronic obstructive pulmonary disease, dementia, liver disease, rheumatologic disease, or malignancy (Supplementary Table 1 presents specific codes). Concomitant pharmacotherapy was assessed from the Prescription Registry as a filled prescription within one year prior to ACS admission. Hypertension was defined from at least two antihypertensive filled prescriptions within one year of ACS admission as done previously.(10) Coronary angiography, (CAG), percutaneous coronary intervention (PCI), and coronary artery bypass grafting (CABG) conducted during ACS admission was assessed and compared between study periods.

Baseline characteristics for patients admitted to hospital with ACS were compared between study periods (1: 1 January -7 May 2017 -2019 2: 1 January-6 May 2020). Categorical variables are presented in counts and percentages and continuous variables are presented with a median and 25 and 75 percentiles. The number of ACS admissions was compared between study groups. For study period 1 (1 January -7 May 2017 -2019), the average number of ACS admissions were computed for the comparison with study period 2 (1 January -6 May 2020) and difference in percentage was computed for the two study periods. The incidence rate of ACS admissions was computed for every week from index date in each study period.

Incidence rates were computed with person years as the denominator and ACS admissions in the nominator. For the computation of the incidence rates the study cohort was split in bands of weeks (7 days) from index date. Incidence rate ratios (IRR) with 95% CI were computed using a Poisson regression analysis for the comparison of every week for the two study periods. In a subgroup analysis, ACS admissions were grouped in 1) STEMI and 2) NSTEMI, UA and AMI unspecified and IRR were computed for the two study periods.

Further, in a sub group analysis, we identified number of ACS admissions by age groups (<60 years and ≥60 years) and by sex for the two study periods.

For a detailed investigation of the period after national lock-down, two additional cohorts were investigated: 1) 12 March -6 May 2017 -2019, and 2) 12 March -6 May 2020.

Baseline characteristics were compared between cohorts and differences in ACS admissions were assessed. Further, incidence rates were computed by ACS subtype for the period after 

An average of 1186 ACS admissions were identified from 12 March -6 May 2017-2019 as compared with 1048 ACS admissions from 12 March -6 May 2020 (11.6% decline). We identified that 59.2% and 54.1% of the ACS patients underwent PCI during ACS admission for the 2017-2019-and 2020-period, respectively. For subgroups of STEMI, NSTEMI, UA, and AMI unspecified, differences of -6.6%, -1.0%, -25.1%, and -29.6% were identified, respectively, for 2020 as compared with the average of 2017-2019. We identified a lower proportion of patients undergoing CAG and PCI during admission for all sub types of ACS (Table 3) . Overall, no difference in baseline characteristics was identified for ACS patients for the two periods (Supplementary Table 2 ). The 30-day mortality was 6.6% and 6.7% for ACS patients in the 2017-2019-and 2020-study period, respectively. In adjusted analysis, no significant difference was found in 30-day mortality, OR=1.03 (95% CI: 0.77-1.37) for patients in 2020 as compared with 2017-2019. No significant difference in 30-day mortality was found between study periods for the different sub types, STEMI p=0.74, NSTEMI p=0.58, UA p=0.71, and AMI unspecified p=0.12 (Table 3 ).

This nationwide study examined the incidence of ACS during national lock-down due to the COVID-19 pandemic. The main finding was that the incidence of ACS admissions was significantly reduced with approximately 30% around the announcement of national lock- for this finding. Speculatively, a higher degree of type 2 AMIs could explain this finding, however from the registries used it was not able to differentiate by type of AMI. The national lock-down in Denmark was announced on March 11, 2020, however from our data a decline in ACS admissions was seen from February 27, 2020 (week 9). We can only speculate on this finding, however awareness in the Danish population of health care systems pushed to the limits in other countries may have led to patients not seeking medical help and fear of virus contraction by contact to the health care system. Further, our data showed that with the slowly reopening of Denmark from April 6 2020, admission rates of ACS have returned towards previous years. This finding may suggest that care is especially needed in patients with ACS around the time of national lock-down and that admission rates normalize when reopening is initiated. This finding may be useful for health care planners and policy makers if a new pandemic or new COVID-19 wave appears. We found no significant reduction in STEMI admissions, however a trend towards a decline in the lock-down period could be interpreted from our results and wide confidence intervals in this subgroup suggest a potential type II-error. In part, patients classified as AMI unspecified may have been patients with STEMI, however we were not able to assess this category in further detail, which is a limitation to the present study. In additional subgroup analyses, we identified that the main result was driven by male patients and patients ≥ 60 years of age. Speculatively and in general terms, male patients may have a tendency to seek medical help less often than female patients. For patients ≥ 60 years, a high concern of contracting COVID-19 when seeking medical help may explain the decline in ACS admissions for this patient group. The Danish health care system has been vastly reorganized at the beginning of the pandemic. Outpatient visits were cancelled or changed to telemedicine if possible and nonvital elective procedures were cancelled, while resources were allocated for emergency rooms and intensive care units. Hence, our findings may be partly explained by both system and patient related factors and we cannot clearly distinguish these from each other.

Little variations were seen in the characteristics of patients admitted with ACS in the lockdown period as compared with previous years. Overall, risk factors for ischemic heart disease were frequent in the population with no difference between groups. Hypothetically, one could have expected that patients with few comorbidities were more present during the lock-down period in order to navigate in a health care system with resources allocated to COVID-19 wards, however this was not seen from the data presented in this study.

The main strength of our study is the large, unselected sample size from a nationwide cohort.

The Danish nationwide health care registries provide a unique opportunity for the investigation of a national lock-down during the COVID-19 pandemic with detailed descriptions on ACS management and mortality.

Our study has some limitations. First, the ACS diagnosis was defined from diagnosis codes and data from electrocardiogram, troponin levels, and coronary angiography were not available. Our data provide knowledge on first-time ACS admissions and not total ACS admissions. Second, from the registries used, we were not able to identify patients with a delayed entry into the health care system and patients with missed STEMI were not assessable. Third, although fluctuations in ACS admission between weeks may be by chance, the changes over time was statistically significant, suggesting that the changes were related to the COVID-19 national lock-down in Denmark. Fourth, prescribed medication was only available up until February 29, 2020, hence we may have underestimated the use of medication in patients admitted with ACS in the most recent time period investigated. Fifth, no data on causes of death were obtained and we were not able to investigate the overall ACS mortality in the calendar periods studied. Sixth, data on ethnicity of included patients were not available. Seventh, patients categorized as AMI unspecified may have withhold patients with either STEMI or NSTEMI and the results of the sub group analysis by type of ACS must be seen in this light.

In conclusion, from nationwide data during a national lock-down due to the COVID-19 pandemic, we identified a significant decrease in ACS admissions with around 30%. Along with the reopening of society, ACS admissions were stabilized at levels equal to previous years. Subgroup analyses showed that the main result mainly was driven by male patients, 

Første dansker testet positiv for COVID-19

Cardiovascular Events during World Cup Soccer

Triggering of Acute Myocardial Infarction by Heavy Physical Exertion --Protection against Triggering by Regular Exertion

Do Episodes of Anger Trigger Myocardial Infarction? A Case-Crossover Analysis in the Stockholm Heart Epidemiology Program (SHEEP): Psychosomatic Medicine

Effect of widespread restrictions on the use of hospital services during an outbreak of severe acute respiratory syndrome

The Danish National Patient Registry: a review of content, data quality, and research potential

The Danish Civil Registration System as a tool in epidemiology

The Danish National Prescription Registry

Positive predictive value of cardiovascular diagnoses in the Danish National Patient Registry: a validation study

Validation of risk stratification schemes for predicting stroke and thromboembolism in patients with atrial fibrillation: nationwide cohort study

R: A language and environment for statistical computing. R Foundation for Statistical Computing

Reduction in ST-Segment Elevation Cardiac Catheterization Laboratory Activations in the United States during COVID-19 Pandemic

Decline of acute coronary syndrome admissions in Austria since the outbreak of COVID-19: the pandemic response causes cardiac collateral damage

Reduced Rate of Hospital Admissions for ACS during Covid-19 Outbreak in Northern Italy

Reduction of hospitalizations for myocardial infarction in Italy in the COVID-19 era

Epub ahead of print

Out-of-Hospital Cardiac Arrest during the Covid-19 Outbreak in Italy

Declines in Hospitalizations for Acute Cardiovascular Conditions During the COVID-19 Pandemic: A Multicenter Tertiary Care Experience

Cardiologists urge heart attack, stroke patients not to delay medical attention over COVID-19 fears

ACS: acute coronary syndrome, IQR: interquartile range, COPD: chronic obstructive pulmonary disease, RASi: renin-angiotensis system inhibitor, ASA: aspirin, ADPi: adenosine-di-phosphate inhibitor, OAC: oral anticoagulant therapy.